Embed Size (px)
Citation preview
This article was downloaded by: [Lib4RI]On: 01 February 2012, At: 02:03Publisher: Taylor & FrancisInforma Ltd Registered in England and Wales Registered Number: 1072954 Registeredoffice: Mortimer House, 37-41 Mortimer Street, London W1T 3JH, UK
International Journal of EnvironmentalHealth ResearchPublication details, including instructions for authors andsubscription information:http://www.tandfonline.com/loi/cije20
A systematic approach to behaviorchange interventions for the waterand sanitation sector in developingcountries: a conceptual model, areview, and a guidelineHans-Joachim Mosler aa Eawag, Swiss Federal Institute of Aquatic Science andTechnology, Ueberlandstrasse 133, 8600, Duebendorf, Switzerland
Available online: 31 Jan 2012
To cite this article: Hans-Joachim Mosler (2012): A systematic approach to behavior changeinterventions for the water and sanitation sector in developing countries: a conceptualmodel, a review, and a guideline, International Journal of Environmental Health Research,DOI:10.1080/09603123.2011.650156
To link to this article: http://dx.doi.org/10.1080/09603123.2011.650156
PLEASE SCROLL DOWN FOR ARTICLE
Full terms and conditions of use: http://www.tandfonline.com/page/terms-and-conditions
This article may be used for research, teaching, and private study purposes. Anysubstantial or systematic reproduction, redistribution, reselling, loan, sub-licensing,systematic supply, or distribution in any form to anyone is expressly forbidden.
The publisher does not give any warranty express or implied or make any representationthat the contents will be complete or accurate or up to date. The accuracy of anyinstructions, formulae, and drug doses should be independently verified with primarysources. The publisher shall not be liable for any loss, actions, claims, proceedings,
demand, or costs or damages whatsoever or howsoever caused arising directly orindirectly in connection with or arising out of the use of this material.
Dow
nloa
ded
by [
Lib
4RI]
at 0
2:03
01
Febr
uary
201
2
A systematic approach to behavior change interventions for the water
and sanitation sector in developing countries: a conceptual model, a
review, and a guideline
Hans-Joachim Mosler*
Eawag, Swiss Federal Institute of Aquatic Science and Technology, Ueberlandstrasse 133, 8600Duebendorf, Switzerland
(Received 19 July 2011; final version received 26 November 2011)
Public health practitioners increasingly agree that it is not enough to providepeople with water and sanitation hardware. Numerous approaches are used totackle the ‘‘software’’ which means to ensure behavior change necessary to comealong with the sanitation hardware. A review of these approaches reveals severalshortcomings, most importantly that they do not provide behavioral changeinterventions which correspond to psychological factors to be changed. Thisarticle presents a sound psychological model, which postulates that for theformation of new habitual behavior, five blocks of factors must be positive withregard to the new behavior: risk factors, attitudinal factors, normative factors,ability factors, and self-regulation factors. Standardized tools for measuring thefactors in face-to-face interviews are presented, and behavioral interventions areprovided for each factor block. A statistical analysis method is presented, whichallows the determination of the improvement potential of each factor.
Keywords: attitude; hygiene; questionnaire; behavior; sanitation
Introduction
Each year 1.5 million children die before their fifth birthday because of diarrhea,nearly all in developing countries (Pruss-Ustun et al. 2008). It is estimated that 88%of these deaths could be prevented by a safe water supply, sanitation, and hygiene.Massive resources have been invested in providing water and sanitation facilities,such as drinking water disinfection technologies, improved toilets, and handwashingstands (Peal et al. 2010). Nevertheless, awareness is growing amongst public healthpractitioners that it is not enough to provide people with facilities – these facilitieswill be useless if used improperly or not at all (Cairncross and Short 2004).Practitioners all over the world report unused or misused toilets (e.g. used as storagerooms), abandoned newly constructed wells, and improperly performed hygiene(Mara et al. 2010; Peal et al. 2010). Evidence suggests that the effectiveness oftechnologies is dependent on the degree of compliance, as Du Preez et al. (2010) andMausezahl et al. (2009) demonstrated in the case of solar water disinfection.Providing populations in developing countries with hygiene, sanitation, and water‘‘hardware’’ must be accompanied by programs that generate behavior change
*Email: [email protected]
International Journal of Environmental Health Research
2012, 1–19, iFirst article
ISSN 0960-3123 print/ISSN 1369-1619 online
� 2012 Taylor & Francis
http://dx.doi.org/10.1080/09603123.2011.650156
http://www.tandfonline.com
Dow
nloa
ded
by [
Lib
4RI]
at 0
2:03
01
Febr
uary
201
2
(Peal et al. 2010). According to Stanton et al. (1992), all efforts to reduce diarrhealmorbidity and mortality require behavioral change.
Behavior is the result of psychological processing of factors within the individual.Behavior change campaigns must take these factors into account. Practitionersshould know which of these factors keeps the target population attached to theunhealthy behavior. For them it is crucial to know which interventions change whichinner factor for conducting successful behavior change campaigns.
There are numerous approaches with which practitioners try to ensure com-pliance by their target population. A recent publication by Peal et al. (2010) itemizesa list of hygiene and sanitation ‘‘software’’ approaches – meaning social inter-ventions that enable change in behavior. These approaches will be reviewed in afollowing chapter in this article, but they can be summarized as lacking (a) asystematic model of behavior-determining factors; (b) a methodology to measurethese behavioral factors; (c) a method to analyze and verify the impact of thesefactors on target population behavior; and (d) instructions for determining thenecessary behavior change techniques based on the preceding analysis. An instru-ment providing all these components would enable behavior change by system-atically applying well-directed behavior change techniques.
This article aims to develop a methodological approach that should allowpurposive behavior change by presenting: (1) a conceptual behavioral model basedon sound psychological evidence and theory, (2) the behavior change techniquescorresponding to the factors to be changed, and (3) an analytical tool for deriving thefactors to be changed on the basis of quantitative data. On the background of thepresented model this article will review how behavioral change is conceptualized,implemented, and measured in the scientific literature and in guidelines on water andsanitation. This review serves to demonstrate potential omissions in procedures thatmight be essential to long-term behavior change. Finally, a general protocol forbehavior change is outlined, followed by concluding remarks. The general aim of thisarticle is to establish a procedure that provides a compelling deduction of behavioralchange interventions from quantitative data.
The RANAS model: r(isk), a(ttitudes), n(orms), a(bilities), and s(elf-regulation)
of behavioral change
The proposed model is divided in four distinctive components: (1) factor blocks, (2)behavioral factors, (3) target behaviors, and (4) behavior change interventionscorresponding to the factor blocks (see Figure 1).
Factor blocks
Five blocks of factors have to be favorable to the new behavior in order for it to takeroot: risk factors, attitudinal factors, normative factors, ability factors, and self-regulation factors (see Figure 1). Several theories of behavioral change describefactors that can neatly be classified into these five factor blocks (see Albarracın et al.2005).
The risk factors block contains all factors that deal with an individual’sunderstanding and awareness of the health risk. Risk perceptions are postulated bythe Health Belief Model (Rosenstock 1974); the Protection Motivation Theory(Floyd et al. 2000); and by the Health Action Process Approach (Schwarzer 2008).
2 H.-J. Mosler
Dow
nloa
ded
by [
Lib
4RI]
at 0
2:03
01
Febr
uary
201
2
The Theory of Planned Behavior (Fishbein and Ajzen 2010) describes attitudinal,normative, and ability factors. Attitudinal factors are those which express a positiveor negative stance toward a behavior. Normative factors represent convictions aboutthe incidence of a behavior and how the social network thinks about the behavior.Ability factors represent aptitudes an individual believes he or she must have inorder to acquire the behavior. Self-regulation factors (Albarracın et al. 2005) areresponsible for the continuance and maintenance of the behavior (as postulated byProchaska and DiClemente 1983).
These theories have been proven useful in explaining and changing differinghealth behaviors (for the Theory of Planned Behavior see Ajzen et al. 2007; for theHealth Action Process Approach see Schwarzer 2008). Several publications haveshown the RANAS Model’s factors influence behavior in the water and sanitationsector in developing countries: for solar water disinfection (SODIS) see Heri andMosler (2008) in Bolivia, and Kraemer and Mosler (2010) in Zimbabwe; for hygienebehavior see Graf et al. (2008) in Kenya; for using arsenic-free deep tube wells seeMosler et al. (2010) in Bangladesh. Analysis of behavioral factors from the pers-pective of health psychology theories in developed countries is a successful means ofpredicting population behavior in the water and sanitation sector of developingcountries.
Behavioral factors
The distinction between perceived vulnerability and perceived severity must be madefor the risk factors. Perceived vulnerability refers to a person’s subjective perceptionof his or her risk of contracting a disease. Perceived severity is a person’s perceptionof the seriousness of the consequences of contracting a disease (Floyd et al. 2000).
Figure 1. The RANAS Model of behavior change.
International Journal of Environmental Health Research 3
Dow
nloa
ded
by [
Lib
4RI]
at 0
2:03
01
Febr
uary
201
2
A person should have an understanding – through environmental, or factual,knowledge – of how she or he could be affected by a disease (Albarracın et al. 2005);e.g. knowing the possibilities for potential pathogen contamination.
Attitudinal factors include instrumental beliefs (outcome expectancies), such asbeliefs about costs in terms of money, time, and effort; and benefits in terms ofsavings, health, or other advantages of the new behavior. Furthermore, attitudeshave an affective component (Trafimow and Sheeran 1998). Affective appraisals(beliefs) are defined as feelings arising when performing or thinking of the behavior.
Several kinds of norms are considered for normative factors. The descriptivenorm refers to perceptions of which behaviors are typically performed by others,whereas the injunctive norm reflects perceptions of which behaviors are typicallyapproved or disapproved of by relatives, friends, or neighbors (Cialdini et al. 2006;Schultz et al. 2007). Included in injunctive norms are institutional norms, which arethe dos and don’ts expressed by recognized authorities such as leaders of villages,tribes, and religious or other institutions. Finally, the personal norm conveys whatan individual personally believes she or he should do (Schwartz 1977). This normmust be taken into account, as it can contradict the other norms.
Ability factors represent the confidence of a person in her or his ability toperform a behavior. One pre-condition, called action knowledge, is that people knowhow to perform the behavior (Frick et al. 2004). Additionally, a positive self-efficacyis needed: the belief in one’s ability to organize and execute the courses of actionrequired to manage prospective situations (Bandura 1997). Two other kinds of self-efficacy are relevant factors in this block. Maintenance (coping) self-efficacy includesbeliefs about one’s ability to deal with barriers that arise during the maintenance ofthe behavior, and recovery self-efficacy describes the experience of failure andrecovery from setbacks (Schwarzer 2008).
Finally, self-regulation factors (Schwarzer 2008), also called self-managementfactors (Bandura 2004; Albarracın et al. 2005), help the person to manage conflictinggoals and distracting cues when intending to implement and maintain a behavior(Gollwitzer and Sheeran 2006). Action control refers to a strategy where the ongoingbehavior is continuously evaluated with regard to a behavioral standard (Schwarzer2008), and action planning represents thoughts about how to set up the behavior byspecifying the when, where, and how of the behavior (Gollwitzer and Sheeran 2006).Coping planning is defined as the presumption of possible barriers and the inventionof ways to overcoming them (Schwarzer 2008). To perform a behavior continuously,the person has to remember the behavior as well as commit to it (Tobias 2009).
Target behaviors
This section will describe the outcomes of the behavioral factors. In addition to thedesirable behavior, competing behaviors must also be considered – e.g. not onlydrinking safe water (Behavior A) but also drinking raw water (Behavior B). Factoroutcomes include not only behavior, but also use, intention, and habit. Behavior inthe water and sanitation sector is mostly related to the use of a technology, e.g. awater source or sanitation system. Therefore, the use of these devices must bemeasured as an outcome of the behavior change process. The intention to perform abehavior is often regarded as a behavior-determining factor resulting from severalbeliefs (see Fishbein and Ajzen 2010). Some psychological theories (e.g. HealthAction Process Approach, Schwarzer 2008) differentiate between a motivational
4 H.-J. Mosler
Dow
nloa
ded
by [
Lib
4RI]
at 0
2:03
01
Febr
uary
201
2
phase, which creates an intention, and a volitional phase, which generates thebehavior. Risk perception, attitudinal factors, and some ability factors contribute tointention building and self-regulation factors, and some ability factors (maintenanceand recovery self-efficacy) contribute to behavior performance. Habits are themost important outcome, as the goal of each behavior change campaign is to builda long-term habitual behavior amongst the majority of the target population. Habitsare formed through an interplay between initial commitment, remembering, andrepeated performance of the behavior. This repetition may be supported by planningfactors (Tobias 2009). Habitual behaviors are performed nearly automatically: withoutany cognitive effort, but based on an intention (Aarts and Dijksterhuis 2000).
Intervention techniques targeting behavioral factors
The intervention techniques corresponding to the factor blocks of the RANASModel should be specified. It is necessary to map which class of techniques will mostlikely change which factors in which factor block. The relationship between factorblocks and interventions is not necessarily a one-to-one correlation. In fact, many ofthe intervention techniques affect more than one factor. However, techniques shouldbe the most efficient way to change the factors to which they are assigned. Theallocation of behavior change techniques to factors of the model is done based onthe literature: the Intervention Mapping Approach of Bartholomew et al. (2006); anextensive meta-analysis of over 350 intervention studies by Albarracın et al. (2005);and a consensus process of expert judges who linked behavior change techniqueswith behavioral factors, undertaken by Michie et al. (2008). Curtis et al. (2009)presented interventions corresponding to their eight postulated motivations, andStanton et al. (1992) relate interventions to behavioral factors in their behavioralintervention framework. Interventions corresponding to the factor blocks elaboratedin the RANAS Model are shown in Table 1.
Information interventions ) risk factors
The risk perceptions block can be by influenced by information interventions. Whenprovided with information, the person should be able to form an understanding ofthe health threat (Stanton et al. 1992). Risk perceptions can be affected with per-sonalized risk messages, which might focus on cumulative risk effects, and by pre-senting qualitative and quantitative examples (Bartholomew et al. 2006, chapter 7).Risk perceptions can also be altered by showing scenario-based risk information(Bartholomew et al. 2006, chapter 7) or with threat-inducing arguments that usefrightening information (Albarracın et al. 2005). Factual knowledge can be increasedby presenting information about the circumstances and possibilities of contracting adisease (Albarracın et al. 2005).
Persuasive interventions ) attitudinal factors
Instrumental beliefs can be changed with persuasive interventions, or strongarguments or peripheral cues as described in the Elaboration Likelihood Model(Petty et al. 2004). Persuasive arguments are those which use causal explanations;explain functionality; present novel and important information; and are of highpositive expectancy value. Persuasive peripheral cues are competence, sympathy,
International Journal of Environmental Health Research 5
Dow
nloa
ded
by [
Lib
4RI]
at 0
2:03
01
Febr
uary
201
2
credibility, famousness, publicity of the source, and length and number of themessage’s arguments (Petty et al. 2004).
Affective beliefs (feelings) may also be changed through persuasive interventions,but these need affective persuasion – presenting the performance of a healthybehavior as joyful, or attaching aversion (e.g. disgust) to an unhealthy behavior(Petty et al. 2004).
Normative interventions ) normative factors
According to Cialdini et al. (2006), injunctive normative messages about a stronglydisapproved of behavior are effective. However, giving the message that an undesiredbehavior is regrettably frequent can be counter-effective, because it emphasizes thedescriptive norm by stating what most people are doing. Rather, the descriptive normcan be changed by highlighting norms of still-infrequent but desired behaviors, or byreducing the ‘‘social pressure’’ to engage in an unfavorable behavior by referring to afavorable injunctive norm (Cialdini et al. 2006). The descriptive norm can be changed
Table 1. Factor blocks, behavioral factors and corresponding interventions in the RANASModel.
Risk factors Behavioral interventions: information interventions
Factual knowledgeVulnerabilitySeverity
Presentation of facts/knowledge transferPersonal risk informationShowing scenariosFear arousal
Attitudinal factors Behavioral interventions: persuasive interventions
Instrumental beliefs Persuasive argumentsPersuasive peripheral cues
Affective beliefs Affective persuasion
Normative factors Behavioral interventions: normative interventions
Descriptive normInjunctive normPersonal norm
Highlighting normsPublic commitmentAnticipated regret
Ability factors Behavioral interventions: infrastructuraland ability interventions
Action knowledge (skills) Knowledge transfer (education)Self-efficacy Guided practice
Facilitating resources (financing)Social helpModeling/vicarious reinforcement
Maintenance (coping) self-efficacy Coping with barriersRecovery self-efficacy Coping with relapse
Self-regulation factors Behavioral interventions: planning interventionsand relapse prevention
Action controlCoping planningRememberingCommitment
Daily routine planningOutcome feedbackContingency managementStimulus controlForming implementation intentionsPrompts
6 H.-J. Mosler
Dow
nloa
ded
by [
Lib
4RI]
at 0
2:03
01
Febr
uary
201
2
with a public commitment by showing that there are people who perform the newbehavior. Anticipated regret – where individuals are encouraged to imagine how theywould feel after they behaved in a way that is not consistent with their personal norms– can reinforce desired behaviors (Bartholomew et al. 2006, chapter 7).
Infrastructural, skill and ability interventions ) ability factors
Infrastructural interventions and ability interventions help individuals gainconfidence in their ability to perform a behavior. Resources in the form of financialor in-kind support may be either given directly to individuals, or coupled to thecondition that the individual must make some effort to access the resources. Addi-tionally, help from neighbors, friends, acquaintances, or relatives can support theperson with material assistance, action knowledge, or verbal social support (Bandura2004). Modeling (seeing someone performing a behavior) and vicarious reinforce-ment (seeing a person being rewarded for a behavior) promote desired behaviors;individuals become aware of their own competency and compare their achievementsto those of others (Bandura 2004). Action knowledge (a particular skill) is enhancedby knowledge transfer (education). Self-efficacy can be improved by guided practice,skill demonstration, instruction, and enactment with feedback (Bartholomew et al.2006; Michie et al. 2008, chapter 7).
Maintenance (coping) self-efficacy can be improved by identifying barriers andplanning solutions to behavior change obstacles (Michie et al. 2008; Schwarzer2008). Coping with relapse will augment recovery self-efficacy. Individuals can copewith relapse by identifying risky situations where they might fall back into the oldbehavior, planning coping responses, and practicing these responses until theybecome automatic (Schwarzer 2008).
Planning interventions and relapse prevention ) self-regulation factors
Planning interventions include implementation intentions, which help individualstranslate goals into actions by preventing them from becoming distracted, helpingthem to avoid falling back into bad habits, or inhibiting their failure to get started(Gollwitzer and Sheeran 2006). Relapse prevention skills can be improved by teachingindividuals to foresee situations with a high risk of behavior lapse (Schwarzer 2008).The individual can deal with conflicting goals of the behavior by considering possiblebarriers and overcoming intervening behaviors (Schwarzer 2008). Coping planninguses stimulus control by removing reminders or cues to engage in old behaviors, andadding cues or reminders to engage in the new behavior. Daily routine planning,outcome feedback, and contingency management foster action control. Daily routineplanning includes a discussion of when and where in the daily routine the newbehavior can be integrated (see Schuz et al. 2007). With outcome feedback, the effectsof the new behavior have to be reported back to the person or the person herselfcontrols for these effects (self-feedback) (Michie et al. 2008). Contingency manage-ment involves increasing the rewards (e.g. financial or material) for positivebehavioral change (Bartholomew et al. 2006, chapter 7). Desirable behaviors canbe supported with prompts set by the individual that trigger the behavior in the rightsituation and help to remember the behavior. An individual can set implementationintentions, formulating when, where, and how to achieve his or her goals (Tobias2009).
International Journal of Environmental Health Research 7
Dow
nloa
ded
by [
Lib
4RI]
at 0
2:03
01
Febr
uary
201
2
The commitment to perform a behavior can be enhanced by making a contractwith the person in which she or he makes an agreement, either privately or in acommunity event, to perform the behavior (Michie et al. 2008).
Communication channels
The behavioral interventions have to be brought to the target population; thus, theadequate method of intervention delivery must be determined, and the communica-tion channels (as defined in Rogers’ Theory of Diffusion of Innovations 1995) haveto be selected. Stanton et al. (1992) included in their intervention framework aselection of communication channels to transport the health-related message to thetarget population. A comprehensive list of communication channels is displayed inUNICEF (1999). For space reasons, a compilation of communication channels usedin the development context is not displayed here; but it can be viewed at http://www.eawag.ch/forschung/siam/schwerpunkte/soziale_systeme/index_EN
Determining the behavioral factors to be changed
Practitioners are advised to first measure the incidence of each of the factors to bechanged in the population, and then analyze the intervention potential of thesefactors. The RANAS Model factors can be measured in a standardized way bydeveloping several questions corresponding to each. The factors and some cor-responding questions are depicted in the Appendix. Table A1 in the appendix pro-vides sample questions that should be adapted for each topic and local condition. Thequestions will be introduced into a standardized questionnaire and arranged in ameaningful sequence. The questionnaire ought to be discussed with people local to theregion to make it understandable to the target population. It has to be translated intothe local language(s) and should then be retranslated to assure that the meanings ofthe questions were translated accurately. An intensive training with the interviewerteam is crucial. Interviewers must understand the questionnaire and be trained for theinterview situation.
Research involving human subjects should be reviewed by an Ethical ReviewBoard, or some similar process, to ensure protection of interview subjects’ rights.Respondents should give informed consent to participate in the survey (and insubsequent interventions). Once this preparatory work is done, the representativesurvey can be conducted.
When practitioners cannot survey the whole target population – e.g. in a smallvillage a random sample should be drawn. Ideally, this sample is drawn by randomlyselecting respondents from a listing of the total population. If such a list is notavailable, as is often the case in developing countries, the sample can be drawn withthe random route method (see Kraemer andMosler 2010). In this method, every thirdhousehold is selected. Intentional selection by the interviewers can thereby beavoided.
The project should manage to control the interviewing, and data quality must beassured. For the analysis it is convenient to introduce the data into a data spreadsheetor data file and processed with a statistical analysis program. First, the behaviorchange practitioner might look at the frequencies and means of the target behavior, aswell as of each factor. To determine the intervention potential of each factor,practitioners will first analyze the size of their improvement reserve, and second,
8 H.-J. Mosler
Dow
nloa
ded
by [
Lib
4RI]
at 0
2:03
01
Febr
uary
201
2
measure their impact on behavior. The improvement reserve (IR) is defined as thedifference between the population mean (Mean) in a factor and the maximumpossible value (Max) of this factor (IR ¼ Max – Mean). To determine the impact ofeach factor compared to the other factors on forming intentions, behaviors, andhabits, regression analyses have to be calculated. In the regression analysis, theB-values express how the size of each factor influences the behavior compared to theothers, provided that the factors are transformed to the same range of values.Nevertheless, one must be careful when interpreting only B-values. Small orinsignificant B-values may occur because nearly all persons in the population have thesame low value in this factor. Consequently, the mean in this factor will be low,meaning that the improvement reserve (IR ¼ Max – Mean) will be large. Finally, tofix the behavior improvement potential (IP) of each factor, the IR has to be multipliedby the B-value derived from the regression analysis (IP ¼ IR 6 B). To take intoaccount the uncertainty of an intervention’s impact on a factor, one can calculate theconfidence interval (CI) of B: the possible minimum (lower limit, LL) and maximumimprovement (upper limit, UL) potential is then calculated as IPmin ¼ IR 6 (B–LL)and as IPmax ¼ IR 6 (B þ UL).
Thus far, this article has outlined how the behavioral factors to be changed mightbe derived in a systematic and comprehensive way. The following chapter will discusshow behavioral factors are drawn from different approaches in the literature aboutbehavioral change in the water and sanitation sector in developing countries.
Behavior change in the water and sanitation in developing countries literature
This section will review the literature on behavioral factors and interventions in orderto show how behavioral change is dealt with within the water and sanitation sector.Using the developed RANAS Model, strengths and shortcomings of the reviewedapproaches will be highlighted, and suggestions will be given on how to improvecommon understanding and knowledge of a systematic approach to behavioralchange. Peer-reviewed publications, reports, and guidelines containing data onbehavior change in the water and sanitation sector are reviewed. Only literature inwhich behavioral factors are explicitly analyzed is considered. Projects which onlyprovide water and sanitation facilities and do not deal with behavioral factors are notregarded as behavioral interventions. These are more ‘‘hardware’’ interventions in thesense of Peal et al. (2010). All three sectors – safe drinking water, sanitation, andhygiene practices – are addressed. They have influences in common (e.g. riskperception), and sector interventions are often combined. The intention of this reviewis not to include all literature on behavior change in the water and sanitation sector.Rather, examples from different streams in behavioral change work are discussed.The review will examine how behavioral factors are generated by differentprocedures, and how behavioral factors are utilized – or are supposed to be utilized –for intervention purposes.
The literature dealing with behavior change is grouped and discussed in fivecategories of approach:
(1) Behavioral factors are derived from layperson psychological knowledge.(2) Behavioral factors are drawn from qualitative research.(3) Behavioral factors are developed out of a participatory formative project
phase.
International Journal of Environmental Health Research 9
Dow
nloa
ded
by [
Lib
4RI]
at 0
2:03
01
Febr
uary
201
2
(4) Behavioral factors are derived from psychological theory.(5) Behavioral factors are founded in psychological theory and measured to
derive interventions.
Behavioral factors are derived from layperson psychological knowledge
In the first group of studies, the authors assume that changing awareness,consciousness, knowledge, or attitudes is sufficient to change behavior. In severalstudies, Quick et al. (2002) used health education and motivational interviewing tochange knowledge, attitudes, and health practices (Thevos et al. 2000; Quick et al.2002). In their studies, they mostly report a significant increase in desirable behavior(Quick et al. 2002). Likewise, Thevos et al. (2000) found an increase in knowledgethat contaminated water causes diarrhea (factual knowledge) and knowledge thatdiarrhea can be avoided by boiling or treating water (action knowledge). Individualsreported a belief that they could avoid diarrhea (self-efficacy). Waterkeyn andCairncross (2005) assumed that Community Health Clubs (CHCs) change normsand beliefs through health education and structured participation, which shouldelicit a focused group dynamic. As qualitative results, the members of the healthclubs reported that they joined the clubs to gain knowledge and to have funparticipating and socializing, and they mentioned a sense of unity with thecommunity. The authors found significant differences between club members and acontrol group in several observed hygiene indicators.
None of the mentioned projects tested whether the assumed factors wereinfluencing behavior – e.g. by comparing the behavior of the group with higherfactor values with the group with lower values. The work of Cairncross et al. (2005)is an exception. In their study about effects of exposure to an awareness campaign onseveral hygiene outcomes, they found that the number of home visits wassignificantly associated with awareness of the need for handwashing before eating(risk perceptions). Furthermore, the recalled health education classes were positivelyassociated with handwashing reported by women.
In conclusion, the discussed behavior change approaches often mention only oneor very few behavioral factors compared to the RANAS Model and the multitude offactors discussed in behavioral psychology (Fishbein and Ajzen 2010). In particular,social norms, ability factors, and self-regulation factors are not or are only sparselyincorporated. For the most part, authors do not measure any change in these factors,and they almost never (with the exception of Cairncross et al. 2005) relate the changein the factors to the change in behavior. Additionally, the mode of operation ofinterventions remains unclear, because the change in the behavioral factors is notanalyzed with regard to defined intervention techniques.
Behavioral factors are derived from qualitative research
This paragraph discusses approaches that extract behavioral factors through quali-tative measures. For example, Jenkins and Curtis (2005) conducted a qualitativeconsumer study to investigate the choice to install latrines in Benin. They con-ceptualized consumer drives as arising from the difference between the ideal state ofpersonal goals and values and the actual state of physical and social conditions. Inforty in-depth interviews, they identified prestige-related (norm) factor drives, well-being related (instrumental) factor drives, and situational-related (ability) factor
10 H.-J. Mosler
Dow
nloa
ded
by [
Lib
4RI]
at 0
2:03
01
Febr
uary
201
2
drives with their associated beliefs and attitudes. Jenkins and Curtis (2005)recognized two prestige drives as motivators: to affiliate with the urban elite, andto gain new experiences and lifestyles. They also identified two well-being drives:family health and safety, and convenience and comfort. Jenkins and Scott’s study(2007) on sanitation demand in Ghana was based on the concept of decisionadoption stages, including a preference, an intention, and a choice stage. Theyapplied a semi-structured questionnaire to 536 households, and compared satis-faction with current place of defecation, top three reasons for building a latrine, andconstraints that would prevent the households from adopting new methods ofsanitation. They concluded that the preference for changing sanitation is mostlydriven by dissatisfaction with current defecation place and an awareness of thebenefits of home toilets. The intention to build a latrine was typically motivated bypositive preference for changing the situation, prioritization, and the absence ofconstraints. The final choice to install a toilet is conditioned by appropriate oppor-tunities to build, related to product choices, cost, building services, soil conditions,and access to good technical information and support.
The weakness of these approaches is the difficulty of deriving behavioral factorsfrom the multitude of answers to open questions. The authors often state that certainfactors somehow emerge from the data, but it is not clear or traceable how this ishappening.
Behavioral factors are developed out of a participatory formative project phase
Most behavior change programs, such as Participatory Hygiene and SanitationTransformation (PHAST; Sawyer et al. 1998); Community-led Total Sanitation(CLTS; Kar and Chambers 2008); or Community Health Clubs (CHC; Waterkeynand Cairncross 2005) rely on behavioral factors elicited through a participatoryformative phase (see also UNICEF 1999). Some of these behavior changeprograms implicitly target distinctive behavioral factors with certain tools theypromote. For example, the glass of water exercise in CLTS (see Kar and Chambers2008, p. 35) tries to use disgust, which is assumed to steer behavior. The CHCapproach explicitly states that ‘‘People change through peer pressure’’ (Waterkeynand Cairncross 2005). All the programs rely mainly on health education andtraining (factual and action knowledge), which means that the target population isgiven information about disease contraction and preventative behaviors. Theseapproaches neglect the fact that behavioral knowledge is not sufficient motivationto perform the behavior, as shown by many psychological theories (Fishbein andAjzen 2010). Furthermore, a participatory process for eliciting behavioral factorshas several dangers: the participants may not speak openly in the presence ofothers; there might be formal and informal leaders present with whom nobodydares to disagree; there might be disrespected minorities who do not dare to sayanything; hidden agendas and former experiences in social life may lead people toexpress opinions in groups which serve their personal goals but have nothing to dowith their real attitudes and behaviors. In short, there will be group processes – associal psychology shows (Turner 1991) – that inhibit the unbiased release ofreasons for behavior in these public situations.
In conclusion, it is doubtful whether socially unbiased reasons for behavior canbe found when using participatory formative research approaches to determinebehavioral factors.
International Journal of Environmental Health Research 11
Dow
nloa
ded
by [
Lib
4RI]
at 0
2:03
01
Febr
uary
201
2
Behavioral factors are derived from psychological theory
This paragraph discusses studies that use psychological theories to build andunderstand behavioral factors. The SaniFOAM approach (Devine 2009) usesseveral psychological theories, and identifies three groups of sanitation behaviordeterminants: opportunity, ability, and motivation (the F of FOAM stands forFocus, meaning the definition of the desired sanitation behaviors and the targetpopulation). Access/availability, product attributes, social norms, and sanctions/enforcement are subsumed in this approach to opportunity. Knowledge, skills andself-efficacy, social support, roles and decisions, and affordability are determinantsof ability. Motivation determinants are attitudes and beliefs, values, emotional/physical/social drivers, competing priorities, intention, and willingness to pay. Theassignment of the determinants to the groups seems arbitrary, and does not coincidewith psychological theory (e.g. social norms in the opportunity block and not in themotivation block). Some of the determinants are very broad and overlap with others(e.g. emotional/physical/social drivers). The behavior change framework should aidin prioritizing interventions and should improve their effectiveness, but unfortu-nately SaniFOAM mentions no behavioral interventions.
Curtis et al. (2009) made a priori predictions about behavioral factors for hygienebehavior based on a conceptual model for which they used modern social psychologyand biological anthropology. The model is comprised of an environmentalcomponent with social, physical, and biological factors, and a brain componentwith planning, motivation, and habit as factors. They conceptualize planning as thepursuit of long-term objectives and habit as automated behaviors and routines.Motivational factors are split into disgust, status, affiliation, attraction, nurture,comfort, and fear. Using focus group discussions in 13 studies in 11 countries, theyworked out key motivations for handwashing, which were disgust, nurture, comfort,and affiliation. They developed practical intervention strategies for each of thesemotivations.
Figueroa and Kincaid (2010) present a very comprehensive review of behavioraltheories. They developed a model of communication for water treatment and safestorage behavior, which includes communication interventions, intermediate out-comes, behavior outcomes, health outcomes, and environmental context. In theintermediate outcomes block, they use individual, household, and communityapproaches to understand health behavior. They mention knowledge, beliefs andattitudes, perceived risk and severity, subjective norms, self-image, emotionalresponse, self-efficacy, empathy and trust, social influence, and personal advocacy asindividual intermediate outcomes. Unfortunately, communication intervention is theonly behavioral intervention mentioned, and they do not make clear which com-munication intervention could change which behavioral determinant.
In sum, some of the approaches in this group are very elaborated and well based inpsychological theory. Nevertheless, an indication of how to measure the factors andthe distinctive allocation of interventions to the behavioral factors is still missing.
Behavioral factors are derived from psychological theory and measured toderive interventions
This paragraph discusses work in which behavioral factors are based on psycho-logical theory and measured accordingly. In some of the presented articles, inter-ventions aimed at changing specific factors are developed.
12 H.-J. Mosler
Dow
nloa
ded
by [
Lib
4RI]
at 0
2:03
01
Febr
uary
201
2
Aunger et al. (2010) state that behavior can be determined by three differentpsychological processes: automatic or habitual responses, motivated or goal-drivenprocesses to satisfy needs, and cognitive causes which reflect conscious concerns.In their study of 802 household interviews about handwashing in Kenya, theyconducted an exploratory factor analysis with the questionnaire items and detectedfour latent variables. The first factor was habit, meaning the automatic responseto handwashing cues and the influence of repeated behavior. The second factor wasthe need to be clean or hygienic. The third factor related to sexual attraction as amotivation for handwashing. The fourth factor was economic constraints, meaningproblems with costs and soap waste. A binominal regression analysis revealed thatonly the habit and economic constraints factors were correlated with observedhandwashing.
Several studies confirm the usefulness of psychological factors in explaining theapplication of solar water disinfection (SODIS) in different countries and underdifferent environmental conditions (urban – rural; wet land – dry land). Heri andMosler (2008) measured the attributes of an innovation derived from the Theory ofDiffusion of Innovations (Rogers 1995), and added descriptive and injunctive norm,promotional effort, and the alternative behavior (boiling) for a survey of 536households in Bolivia. Using a regression analysis, they showed that these factorsexplained the intention to use, and the consumption of, SODIS to a high degree(intention 52%; behavior 69%). Kraemer and Mosler (2010) surveyed the use ofSODIS in 878 households in slums of Harare (Zimbabwe). They relied on theRANAS Model factors, and additionally a self-persuasion factor (talking about aninnovation convinces the person herself) from persuasion research. Using regressionanalysis, they could successfully determine uptake of SODIS. Finally, Graf et al.(2008) analyzed hygiene behavior and SODIS uptake in 500 households in theKibera slum in Nairobi (Kenya). They measured factors like perceived risk andseverity, lay ideas of diarrhea causes, biomedical knowledge, action knowledge,belief in importance of clean water, and social norms. Using regression analysis, theycould explain SODIS uptake and hygiene behavior with these factors in asatisfactory manner. Mosler et al. (2010) interviewed 222 households in Bangladeshto work out the factors influencing the use of arsenic-free deep tube wells. Theyintroduced personal, social, and situational factors in separated regression analysis,and then conducted an analysis containing all significant factors. With thisapproach, they could explain a large amount of behavioral variance (59%).
In all the investigations mentioned here, behavioral factors are used quite suc-cessfully to explain the respective behavior. From these results, interventionswere proposed to change the statistically significant factors; but tests of these inter-ventions’ modes of action are not reported. Reliable testing of behavioral interventionsremains a gap in behavior change research. Until we know which interventions changewhich behavioral factors, behavior change will remain a blind trial-and-errorprocedure.
The eight-step protocol for behavior change
The short review revealed that in nearly all cases, we do not know whichinterventions are successful, or how they work. This makes knowledge accumulationimpossible with regards to the very important topic of behavior change in the waterand sanitation sector. If we agree that it is necessary to know how interventions
International Journal of Environmental Health Research 13
Dow
nloa
ded
by [
Lib
4RI]
at 0
2:03
01
Febr
uary
201
2
change behavioral factors in order to conduct behavior change in an elaborated andsystematic way, then we have to develop a research strategy that will provide us withthis know-how on a reliable basis. An eight-step general protocol for conductingsystematic behavior change is presented as follows.
Defining target behavior and population
The behavior to be changed and the target population must be defined exactly. Inline with the SaniFOAM approach (Devine 2009) practitioners must determinewhich and whose behaviors require improvement.
Formative research
Research should be conducted to get a first impression about the favoring andhindering conditions of the behavior in question. Different methods were developedfor this task as the Methodology for Participatory Assessments (MPA, van Wijk andPostma 2003).
Identifying behavioral factors
The relevant behavioral factors must be identified using psychological theory.Behavioral factors are sufficiently defined in the RANASModel mentioned here, andcan serve as a blueprint.
Measuring behavioral factors
The behavioral factors – together with intention, habits, and the performance of thebehavior – have to be measured with a questionnaire or by observation. Sampleitems are presented in the appendix of this article, but they must be adapted to thelocal cultural context. Face-to-face interviews have to be conducted with a repre-sentative and randomly selected sample. If financial resources are small, practitionersshould interview at least 30–50 randomly selected households. Practitioners shouldnever rely on the opinions of only a few persons.
Defining target factors
The relevant factors that actually steer the behavior should be identified usingstatistical analysis. Analysis of factor frequency indicates which factors are to beimproved and which are already at a behavior-favoring level. Additionally, regres-sion analyses can be calculated which give out the relative weight of each factor’sinfluence on behavior, intention and habit (see Heri and Mosler 2008; Kraemer andMosler 2010).
Defining interventions
Interventions to change the target behavioral factors should be defined. This can bedone within the framework of the RANAS Model (see Table 1) or, e.g. with theintervention mapping approach of Bartholomew et al. (2006). The interventions haveto be designed and implemented.
14 H.-J. Mosler
Dow
nloa
ded
by [
Lib
4RI]
at 0
2:03
01
Febr
uary
201
2
Evaluating interventions
The effects and the effectiveness of the interventions must be evaluated byconducting a panel survey with the same sample. The main task is to measure thetarget behavior. Whenever possible, this should be done by direct observation toavoid reporting biases, although observation might introduce a reactivity bias. Inaddition to measuring the behavioral factors, practitioners must detect which inter-vention the subjects report having received (intervention check). If the effects ofthe interventions are not satisfactory, the procedure should be repeated from thefourth step.
Evaluating sustainability
To determine the lasting effects of the interventions, practitioners must measurebehavior and behavioral factors 6–12 months after the last intervention to assesssustainable change.
Studies that report carefully on all eight steps will contribute to the cumu-lative scientific knowledge about behavior change in the water and sanitationsector.
Conclusions
A general model of behavior change for water and sanitation was presented, whichconnected behavioral factors from psychological theory with behavioral interventionsfound by the literature to have an impact on these factors. The RANAS Modelcontains five blocks of factors, which are conceptualized to be the main drivers ofbehavior and habit formation: risk, attitudinal, normative, ability, and self-regulationfactors. The model is generic in the sense of Stanton et al. (1992), as it can be appliedto any health behavior, but it has to be adapted to local environmental, social, andcultural conditions. It should serve as a blueprint, which guides successful behaviorchange campaigns by defining the relevant factors and corresponding behavioralchange techniques. In the end it has to be shown that intervention programs followingthe RANAS Model are more successful in changing behavior than using otherapproaches.
The RANAS Model focuses on changes that can be realized by the householdsthemselves; it makes no statements about changes on the meso- and macro-level,such as institutions, the economic or political system, or even changes in theenvironment. However, all these meso- and macro-systems need to influence factorsin the individual, otherwise they will not have any impact on his or her behavior. Themodel is valuable in contexts where households are able to change the conditions oftheir daily life on their own and do not depend on help from outside – e.g. fromgovernmental institutions. If public health practitioners choose to work on theindividual level the presented model can serve as a comprehensive approach tobehavior change.
The proposed eight-step general procedure for conducting behavior changecampaigns outlines steps to induce behavior change in the water and sanitationsector in a systematic way, so that cumulative scientific knowledge can be gained.This procedure should make it possible to reveal the interventions’ rules of action.Practitioners can gain a growing understanding of how to conduct behavior changecampaigns purposefully and with the largest possible effects.
International Journal of Environmental Health Research 15
Dow
nloa
ded
by [
Lib
4RI]
at 0
2:03
01
Febr
uary
201
2
Acknowledgments
I thank Alexandra Huber, Jennifer Inauen, and Rick Johnston for their valuable input intothis work. Alexandra Huber and Jennifer Inauen compiled the basis for the questionnaire.Robert Tobias provided the idea for the improvement potential calculation.
References
Aarts H, Dijksterhuis A. 2000. Habits as knowledge structures. Automaticity in goal-directedbehavior. J Pers Soc Psychol. 78:53–63.
Ajzen I, Albarracın D, Hornik R. (Eds.). 2007. Prediction and change of health behavior:applying the reasoned action approach. New Jersey: Lawrence Erlbaum.
Albarracın D, Gilette JC, Earl AN, Glasman LR, Durantini MR, Ho M-H. 2005. A test ofmajor assumptions about behavior change: a comprehensive look at the effects of passiveand active HIV-prevention interventions since the beginning of the epidemic. Psychol Bull.131(6):856–897.
Armitage CJ. 2005. Can the theory of planned behavior predict the maintenance of physicalactivity? Health Psychol. 24(3):235–245.
Aunger R, Schmidt W-P, Ranpura A, Coombes Y, Maina PM, Matiko CN, Curtis V. 2010.Three kinds of psychological determinants for hand-washing behaviour in Kenya. Soc SciMed. 70:383–391.
Bandura A. 1997. Self-efficacy: the exercise of control. New York: Freeman.Bandura A. 2004. Health promotion by social cognitive means. Health Educ Behav.
31(2):143–164.Bartholomew LK, Parcel GS, Kok G, Gottlieb NH. (Eds.). 2006. Planning health promotion
programs: an intervention mapping approach. San Francisco (CA): Jossey-Bass.Cairncross S, Shordt K. 2004. It does last! Some findings from a multi-country study of
hygiene sustainability. Waterlines. 22:3–5.Cairncross S, Shordt K, Zacharia S, Govindan BK. 2005. What causes sustainable changes in
hygiene behaviour? A cross-sectional study from Kerala, India. Soc Sci Med. 61:2212–2220.
Cialdini RB, Demaine LJ, Sagarin BJ, Barrett DW, Rhoads K, Winter PL. 2006. Managingsocial norms for persuasive impact. Soc Influence. 1(1):3–15.
Curtis V, Danquah LO, Aunger RV. 2009. Planned, motivated and habitual hygienebehaviour: an eleven country review. Health Educ Res. 24(4):655–673.
DeLeon IG, Fuqua RW. 1995. The effects of public commitment and group feedback oncurbside recycling. Environ Behav. 27:233–250.
Devine J. 2009. Introducing SaniFOAM: a framework to analyze sanitation behaviors todesign effective sanitation programs. Global Scaling up Sanitation Project. Water andSanitation Program.
Du Preez M, McGuigan KG, Conroy RM. 2010. Solar disinfection of drinking water in theprevention of dysentery in South African children aged under 5 years: the role ofparticipant motivation. Environ Sci Technol. 44(22):8744–8749.
Figueroa ME, Kincaid DL. 2010. Social, cultural and behavioral correlates of householdwater treatment and storage. Baltimore: John Hopkins Bloomberg School of PublicHealth, Center for Communication Programs.
Fishbein M, Ajzen I. 2010. Predicting and changing behavior: the reasoned action approach.New York: Psychology Press (Taylor & Francis).
Floyd DL, Prentice-Dunn S, Rogers RW. 2000. A meta-analysis of research on ProtectionMotivation Theory. J Appl Soc Psychol. 30(2):407–429.
Frick J, Kaiser FG, Wilson M. 2004. Environmental knowledge and conservation behavior:exploring prevalence and structure in a representative sample. Pers Individ Dif. 37:1597–1613.
Gollwitzer PM, Sheeran P. 2006. Implementation intentions and goal achievement: a meta-analysis of effects and processes. Adv Exp Soc Psychol. 38:249–268.
Graf J, Meierhofer R, Wegelin M, Mosler H-J. 2008. Water disinfection and hygienebehaviour in an urban slum in Kenya: impact on childhood diarrhoea and influence ofbeliefs. Int J Environ Health Res. 18(5):335–355.
16 H.-J. Mosler
Dow
nloa
ded
by [
Lib
4RI]
at 0
2:03
01
Febr
uary
201
2
Harland P, Staats H, Wilke HAM. 2007. Situational and personality factors as direct orpersonal norm mediated predictors of pro-environmental behavior: questions derivedfrom norm-activation theory. Basic Appl Soc Psychol. 29(4):323–334.
Heri S, Mosler H-J. 2008. Factors affecting the diffusion of solar water disinfection: a fieldstudy in Bolivia. Health Educ Behav. 35(4):541–560.
Jenkins MW, Curtis V. 2005. Achieving the ‘good life’: why some people want latrines in ruralBenin. Soc Sci Med. 61:2446–2459.
Jenkins MW, Scott B. 2007. Behavioral indicators of household decision-making and demandfor sanitation and potential gains from social marketing in Ghana. Soc Sci Med. 64:2427–2442.
Kar K, Chambers R. 2008. Handbook on community led total sanitation. Plan and Instituteof Development Studies, University of Sussex, UK; [cited 2011 May]. Available from:http://www.communityledtotalsanitation.org/resource/handbook-community-led-total-sanitation.
Kraemer SM, Mosler H-J. 2010. Persuasion factors influencing the decision to use sustainablehousehold water treatment. Int J Environ Health Res. 20:61–79.
Mara D, Lane J, Scott B, Trouba D. 2010. Sanitation and health. PLoS Med. 7(11):e1000363.Marsh RL, Hicks JL, Landau JD. 1998. An investigation of everyday prospective memory.
Mem Cognit. 26:633–643.Mausezahl D, Christen A, Pacheco GD, Tellez FA, Iriarte M, Zapata ME. 2009. Solar
drinking water disinfection (SODIS) to reduce childhood diarrhoea in rural Bolivia: acluster-randomized, controlled trial. PLOS Med. 6(8):e1000125.
Michie S, Johnston M, Francis J, Hardeman W, Eccles M. 2008. From theory to intervention:mapping theoretically derived behavioural determinants to behaviour change techniques.Appl Psychol: Int Rev. 57(4):660–680.
Mosler H-J, Blochliger OR, Inauen J. 2010. Personal, social, and situational factors in-fluencing the consumption of drinking water from arsenic-safe deep tubewells inBangladesh. J Environ Manag. 91:1316–1323.
Orbell S, Blair C, Sherlock K, Conner M. 2001. The theory of planned behavior and ecstasyuse: roles for habit and perceived control over taking versus obtaining substances. J ApplSoc Psychol. 31:31–47.
Orbell S, Lidierth CJ, Geeraert N, Uller C, Uskul AK, Kyriakaki M. 2009. Social-cognitivebeliefs, alcohol, and tobacco use: a prospective community study of change following aban on smoking in public places. Health Psychol. 28:753–661.
Park HS, Smith SW. 2007. Distinctiveness and influence of subjective norms, personaldescriptive and injunctive norms, and societal descriptive and injunctive norms onbehavioral intent: a case of two behaviors critical to organ donation. Human CommunRes. 33:194–218.
Peal A, Evans B, van der Voorden C. 2010. Hygiene and sanitation software. An overview ofapproaches. Geneva (Switzerland): Water Supply & Sanitation Collaborative Council.
Petty RE, Rucker DD, Bizer GY, Cacioppo JT. 2004. The elaboration likelihood model ofpersuasion. In JS. Seiter & RH. Gass (Eds.), Persuasion, social influence, and compliancegaining. Boston: Pearson. p. 65–90.
Prochaska JO, DiClemente CC. 1983. Stages and processes of self-change of smoking: towardan integrative model of change. J Consult Clin Psychol. 51(3):390–395.
Pruss-Ustun A, Bos R, Gore F, Bartram J. 2008. Safer water, better health: costs, benefits andsustainability of interventions to protect and promote health. Geneva: World HealthOrganization.
Quick RE, Kimura A, Thevos A, Tembo MK, Shamputa I, Hutwagner L, Mintz E. 2002.Diarrhea prevention through household-level water disinfection and safe storage inZambia. Am J Trop Med Hyg. 66(5):584–589.
Rogers EM. 1995. Diffusion of innovations (4th ed.). New York: Free Press.Rosenstock I. 1974. Historical origins of the health belief model. Health Educ Monogr.
2(4):175–183.Sawyer R, Simpson-Hebert M, Wood S. 1998. PHAST step-by-step guide: a participatory
approach for the control of diarrhoeal disease [Internet]. Geneva (Switzerland): WSP/World Bank, UNDP and WHO; [cited 2011 May]. Available from: http://www.who.int/water_sanitation_health/hygiene.
International Journal of Environmental Health Research 17
Dow
nloa
ded
by [
Lib
4RI]
at 0
2:03
01
Febr
uary
201
2
Schuz B, Sniehotta FF, Schwarzer R. 2007. Stage-specific effects of an action controlintervention on dental flossing. Health Educ Res. 22:332–341.
Schultz PW, Nolan JM, Cialdini RB, Goldstein NJ, Griskevicius V. 2007. The constructive,destructive, and reconstructive power of social norms. Psychol Sci. 18(5):429–434.
Schwartz SH. 1977. Normative influence on altruism. In L. Berkowitz (Ed.), Advances inexperimental social psychology, Vol. 10, 221–279. New York: Academic Press.
Schwarzer R. 2008. Modeling health behavior change: how to predict and modify theadoption and maintenance of health behaviors. Appl Psychol: Int Rev. 57(1):1–29.
Smith JR, Terry DJ, Manstead ASR, Louis, WR, Kotterman D, Wolfs J. 2008. The attitude-behavior relationship in consumer conduct: the role of norms, past behavior, and self-identity. J Soc Psychol. 148:311–333.
Stanton B, Black R, Engle P, Peltok G. 1992. Theory-driven behavioral intervention researchfor the control of diarrheal diseases. Soc Sci Med. 35:1405–1420.
Thevos AK, Quick RE, Yanduli V. 2000. Motivational Interviewing enhances the adoption ofwater disinfection practices in Zambia. Health Promot Int. 15(3):207–214.
Tobias R. 2009. Changing behavior by memory aids: a social psychological model ofprospective memory and habit development tested with dynamic field data. Psychol Rev.116:408–438.
Trafimow D, Sheeran P. 1998. Some tests of the distinction between cognitive and affectivebeliefs. J Exp Soc Psychol. 34(4):378–397.
Turner JC. 1991. Social influence. Bristol: Open University Press.UNICEF. 1999. A manual on communication for water supply and environmental sanitation
programmes. Water, Environment and Sanitation Technical Guidelines, 7 [cited 2011May]. Available from: http://www.unicef.org/wash/files/com_e.pdf.
van Wijk C, Postma L. 2003. MPA: a new methodology for participatory monitoring.Waterlines. 22:1.
Waterkeyn J, Cairncross S. 2005. Creating demand for sanitation and hygiene throughCommunity Health Clubs: a cost-effective intervention in two districts in Zimbabwe. SocSci Med. 61:1958–1970.
Appendix
Table A1. Example of a questionnaire about drinking raw/unsafe water versus disinfectedwater.
Factor Item example (in [ ] the response scales)
Vulnerability(Orbell et al. 2009)
How high or low are the chances that you contract diarrheawhen drinking unsafe water?[–4 ¼ very low . . . . . 4 ¼ very high]
Severity(Orbell et al. 2009)
If you contracted diarrhea, how severely would that impactyour life?[0 ¼ not severe . . . . 4 ¼ very severely]
Factual Knowledge(Frick et al. 2004)
How do you think that you get diarrhea when drinking rawwater? [Open-ended]
Instrumental beliefs(Fishbein and Ajzen 2010)
Do you think that using disinfected water istime-consuming (expensive/healthy/effortful)?[0 ¼ not at all expensive/healthy/effortful . . . . 4 ¼ veryexpensive/healthy/effortful]
Affective beliefs(Trafimow andSheeran 1998)
How much do you like or dislike drinkingdisinfected water?[–4 ¼ I dislike it very much . . . . 4 ¼ I rather like it]
Personal norm(Harland et al. 2007)
Do you feel a strong personal obligation to consumedisinfected water?[–4 ¼ I strongly disagree . . . . 4 ¼ I strongly agree]
(continued)
18 H.-J. Mosler
Dow
nloa
ded
by [
Lib
4RI]
at 0
2:03
01
Febr
uary
201
2
Table A1. (Continued).
Factor Item example (in [ ] the response scales)
Descriptive norm(Smith et al. 2008)
How many of your relatives drink disinfected water?[0 ¼ (Almost) nobody (0%) . . . . 4 ¼ (Almost)all (100%)]
Injunctive norm(Park and Smith 2007)
Do you think that, overall, people who are important toyou rather approve or disapprove that you drinkdisinfected water?[–4 ¼ nearly all disapprove . . . . 4 ¼ nearly all approve]
Action knowledge(Frick et al. 2004)
What can be done to avoid diarrhea and its harmful effects?[Multiple choice answers, for each 0 ¼ answer waswrong; 1 ¼ answer was right]
Self-efficacy(Armitage 2005)
Are you sure that you can consume as much disinfectedwater as you need within the next month?[0 ¼ very unsure . . . . 4 ¼ very sure]
Maintenance self-efficacy(Schwarzer 2008)
How confident are you that you can consume as muchdisinfected water as you want, even if your relativescontinue to consume raw water?[0 ¼ not confident at all . . . . 4 ¼ very confident]
Recovery self-efficacy(Schwarzer 2008)
Imagine you have stopped drinking disinfected water forseveral days. How confident are you to start drinkingdisinfected water again?[0 ¼ not confident at all . . . . 4 ¼ very confident]
Action Control (Planning)(Schwarzer 2008)
Do you have a detailed plan for when during the day tostart to disinfect water?[0 ¼ No detailed plan at all . . . . 4 ¼ Very detailed plan]
Coping Planning(Schwarzer 2008)
Have you made a detailed plan regarding what to do whenyou are hindered to disinfect your drinking water?[0 ¼ No detailed plan at all . . . . 4 ¼ Very detailed plan]
Remembering/Forgetting(Marsh et al. 1998)
How often does it happen that you forget to disinfect yourdrinking water?[0 ¼ almost never . . . . 3 ¼ almost always]
Commitment(DeLeon and Fuqua 1995)
How committed do you feel to drink disinfected water?[0 ¼ not at all. . . . 4 ¼ very much]
Intention(Fishbein and Ajzen 2010)
How strongly do you intend to always drink disinfectedwater? [0 ¼ not at all . . .. . ..4 ¼ very strongly]
Behavior(Fishbein and Ajzen 2010)
Percent of disinfected drinking water of total daily waterconsumption. [%]
Habit(Orbell et al. 2001)
Do you go to disinfect water automatically?[0 ¼ No, not at all automatically . . . . 4 ¼ veryautomatically]
Note: Operationalization of the behavioral factors with references in brackets.
International Journal of Environmental Health Research 19
Dow
nloa
ded
by [
Lib
4RI]
at 0
2:03
01
Febr
uary
201
2
